How Does Point Per Possession Work?

[JingleJangleJingle]

What qualifies on getting scored on because i was looking at the amount PPP Tim Duncan gives up and it's around 0.90 while Larry Sanders is around 0.73...Does this mean Larry Sanders is a better 1 on 1 defender or what?

[FuzzyLumpkins]

It means that when he is on the court they take the number of points scored and divide it by the number of possessions. I didn't know Sanders' was so low. That is phenomenal.

Individual defensive metrics are nearly impossible because defense is a team effort. Offense is too but only one person scores. With people helping and rotating and what not defense is not as cut and dry.

[Pop]

Sanders could win both MIP and DPOY.

[JingleJangleJingle]

It means that when he is on the court they take the number of points scored and divide it by the number of possessions. I didn't know Sanders' was so low. That is phenomenal.

Individual defensive metrics are nearly impossible because defense is a team effort. Offense is too but only one person scores. With people helping and rotating and what not defense is not as cut and dry.

i just wondered how they tracked it...if a player blows by Gary Neal and Duncan rotates to help and the guy scores is Gary Neal charged with letting that guy score or is Duncan?

[spurraider21]

i just wondered how they tracked it...if a player blows by Gary Neal and Duncan rotates to help and the guy scores is Gary Neal charged with letting that guy score or is Duncan?

much like +/- stats, i think it goes to everybody on the court at the time. there are different metrics that can analyze defense on post ups, isolations, etc, but those aren't official NBA stats and so aren't always completely reliable

[Mal]

Sanders could win both MIP and DPOY.

You`ve gotta be kidding me.

[mrjap2x]

You`ve gotta be kidding me.
Considering he leads the blocks and his 97.9 defensive rating is quite near duncan, DPOY doesn't seem unlikely. At the very least he is way better than ibaka which has just 99.6 d rating.

[Mal]

Considering he leads the blocks and his 97.9 defensive rating is quite near duncan, DPOY doesn't seem unlikely. At the very least he is way better than ibaka which has just 99.6 d rating.

He`ll get votes for sure, especially from Spurs haters, but Duncan`s case is rock solid this year. Ibaka, Noah, Howard - they all have let down years. And those are big names, who shouldnt be even considered. If Duncan`s ever going to receive DPOTY, it is going to be this year, and Larry Sanders wouldnt take it from him.

[TJastal]

He`ll get votes for sure, especially from Spurs haters, but Duncan`s case is rock solid this year. Ibaka, Noah, Howard - they all have let down years. And those are big names, who shouldnt be even considered. If Duncan`s ever going to receive DPOTY, it is going to be this year, and Larry Sanders wouldnt take it from him.

Add Hibbert to the mix.... he is having an off year offensively but he is anchoring one of the best defenses in the league.

[mrjap2x]

Though I don't think memphis guys will win it, kinda similar to bowen duncan pairing where they will take votes from each other. I think at least one should be in consideration.

Haven't watch them play, and I'd like to ask which of those are most responsible for the defense. Is it conley, allen or gasol.

[HarlemHeat37]

Sanders is actually a very good overall defender, this isn't Marcus Camby or Serge Ibaka, tbh..

I have him as MIP and DPOY is possible, too..

[Johnny RIngo]

Harlem's right. Unlike Ibaka(The Thunder are actually a better defensive team when he's NOT on the floor) Sanders is actually a very good defender. The advanced metrics support this.

[cd021]

You`ve gotta be kidding me.

Why is that so crazy. He seems to be the favorite in both categories.

[Mal]

Why is that so crazy. He seems to be the favorite in both categories.

It is crazy. He`s got most blocks per game in the league. That`s all.

[cd021]

It is crazy. He`s got most blocks per game in the league. That`s all.

He was glued to the bench last season now he is one of the best defenders in the NBA. Leads the league in blocks and Defensive rating. Can you recall such an improvement for a seldom used lotto pick 3 years into their career?

Larry sanders-9.2 ppg (+5.2), Bpg 3.1 (+1.7) rpg 9.2 (+6.1) FG% 50.9 (+5.2%) FT% (+13.8%) P.E.R -18.2 (+5.9)

Per 36- 12.3 ppg, 12.4 rpg, 4.1 bpg, and also turning the ball over much less (down .7 per 36 from last season).

He is clearly the front runner in both categories. That is all.

[jestersmash]

i just wondered how they tracked it...if a player blows by Gary Neal and Duncan rotates to help and the guy scores is Gary Neal charged with letting that guy score or is Duncan?

Both Gary Neal and Duncan are hurt in terms of PPP in this case.

Likewise, if Gary Neal lets his man blow by and Tim makes a spectacular block, both Gary and Tim (and the 3 remaining players on the court at the time) benefit from Tim blocking that shot when in fact the defense on that possession is probably due mostly to Tim.

In this scenario, Gary shows poor defense by letting his man blow by him. Tim shows great defense by blocking the shot near the rim. Every Spur on the court (including Gary and Tim) benefit in the "gives up X points per possession" stat.

I mean, this is why +/- or even rate adjusted +/- (RAPM) is an awful individual stat to look at. RAPM, or points allowed per possession, is best looked at in 5 man units.

If you look at the points allowed per possession for a 5 man unit of Duncan, Parker, Ginobili, Leonard, Green versus points allowed per possession of a 5 man unit of Duncan, Parker, Ginobili, Leonard, and Neal, that difference can be useful in determining the relative difference in defense between Green and Neal.

Looking at overall points allowed per possession of any one individual player is always subject to confounding. Comparing 5 man units as I did above can greatly reduce the effects of confounding.

Edit: But again, the 5 man unit comparisons still don't solve the problem of confounding entirely. In fact, this sort of comparison may not even put a significant dent in the problem of confounding. You'll still have players on court benefiting from the defensive work of others. Likewise, you'll still have players on court penalized for points scored when their individual defense (qualitatively) wasn't at fault.

That's why the "eye test" is probably more important for judging defensive prowess than any one individual stat (BPG, defensive/offensive rebounds) or advanced stat (PPP allowed, etc.)